Early Rumour Detection on Social Media

This is a research project that work towards early rumour detection. Our Rumour detection task is to identify pieces of information on social media that is need to be verified. Manual annotation of large-scale and noisy social media data for rumors is highly labor-intensive, time-consuming and requires special skills and insight to a specific event. Existing rumor dataset are in relatively small size and suffer high class imbalance. We are focusing on three problems: 1) the challenge of data scarcity; 2) modeling algorithm(temporal, propagation) with limited information for early rumor detection at message-level; 3) standardising evaluation and dataset.